CUSTOMER

CHURN RATE
USING MACHINE
IEOR (PE-206) Mid Term Project
LEARNING
-Dhruv Singhal (2K20/PE/25)
-Nipun Chopra (2K20/PE/39)
 ABSTRACT
MACHINE LEARNING
In the last five years there has been a
leap in the visibility of machine
learning, with computational algorithms
now carrying out tasks that a decade ago
seemed far off or even impossible.
Examples range from the vast
improvements in speech recognition and
machine translation of natural language
to medical diagnostics.
2
CUSTOMER CHURNING
Customer churn is defined as when customers
or subscribers discontinue doing business with
a firm or service.
Customers in the telecom industry can choose
from a variety of service providers and
actively switch from one to the next. The
telecommunications business has an annual
churn rate of 15-25 percent in this highly
competitive market and that is why there is a
urgent need to find new and effective ways to
reduce the churn rate.
 ALL ABOUT DATA
The backbone of our project is our Data, we will try to collect the data mostly available to us via
public datasets like UCL machine learning library, Kaggle, US and Indian Govt Public Datasets.
After collection, Data cleaning and primary analysis will be done to set up the premise for our
project, the analysis will include graphical representations and a dashboard to monitor live growth
of the churn rate.
We will explore the data and try to answer some questions like:
• What's the % of Churn Customers and customers that keep in with the active
services?
• Is there any patterns in Churn Customers based on the gender?
• Is there any patterns/preference in Churn Customers based on the type of service
provided?
• What's the most profitable service types?
• Which features and services are most profitable?
• Many more questions that will arise during the analysis

3
DAT
 OUR
APPROACH
Our approach is based on finding new factors and using statistics to pinpoint to
a specific issue whilst suggesting new ways to reduce churn rate by using
comprehensive data of users. This Project provides a short overview of how to
use machine learning to build data-driven models in business. The process of
machine learning is broken down into five stages:

▪ (1) Formulating a problem to model

▪ (2) Collecting and curating training data to inform the model

▪ (3) Choosing an architecture with which to represent the model

▪ (4) Designing a loss function to assess the performance of the model

▪ (5) Selecting and implementing an optimization algorithm to train the

model

At each stage, we discuss how prior physical knowledge may be embedding

into the process, with specific examples.
4
 “Why Are Companies Using
Machine Learning On Churn
Rate”

“
Individualized customer retention is tough
because most firms have a large number of
customers and can't afford to devote much
time to each of them. The costs would be
too great, outweighing the additional
revenue. However, if a corporation could
forecast which customers are likely to leave
ahead of time, it could focus customer
retention efforts only on these "high risk"
clients. The ultimate goal is to expand its
coverage area and retrieve more customers
loyalty. The core to succeed in this market
lies in the customer itself.

5
 “
To reduce customer churn, telecom companies
need to predict which customers are at high risk
of churn.
▪ To detect early signs of potential churn, one must first
develop a holistic view of the customers and their interactions
across numerous channels, including store/branch visits,
product purchase histories, customer service calls, Web-based
transactions, and social media interactions, to mention a few.
▪ As a result, by addressing churn, these businesses may not
only preserve their market position, but also grow and thrive.
More customers they have in their network, the lower the cost
of initiation and the larger the profit. As a result, the
company's key focus for success is reducing client attrition
and implementing effective retention strategy.

6
 DATA COLLECTION
▪ Using dataset websites like Kaggle we acquired the data of
a telecom company’s Customer Churn data.
▪ This is a small sample of the data
▪ The complete data has 7032 rows and 21 columns

7
 Loading Libraries And Data
▪ To perform Analysis on the Data we have to import the data
on the Jupyter Notebook (Python IDE)
▪ To perform Analysis and Understand the data, we have to
add libraries in Jupyter Notebook so we can code our way
through with Python.
▪ These are the Libraries we had to use to Analyse the data.

8
 Understanding the Data
▪ Let us first start with exploring our data set, to better
understand the patterns in the data and potentially form
some hypothesis. First we will look at the distribution of
individual variables and then slice and dice our data for any
interesting trends.
▪ We will be understanding the:
▫ Demographics
▫ Customer Account Information
▫ Services Provided
▫ Monthly and Yearly Charges
▫ Churn Rate
9
 Demographics
▪ Let us first understand the gender, age range, patner and
dependent status of the customers

1. Gender Distribution - About half of the customers in our data set are
male while the other half are female.

2. % Senior Citizens - There are only 16% of the customers who are
senior citizens. Thus most of our customers in the data are younger
people.

10
 Customer Account Information
▪ Let us now look at the tenure, contract
1. Tenure- After looking at the below histogram we can see that a
lot of customers have been with the telecom company for just a
month, while quite a many are there for about 72 months. This
could be potentially because different customers have different
contracts. Thus based on the contract they are into it could be
more/less easier for the customers to stay/leave the telecom
company.

11
 Customer Account Information
2. Contract- To understand the graph below, lets first look at the # of customers by
different contracts.

12
 Services Used
▪ Let us now look at the distribution of various services used
by customers

13